Curr Genet (1997) 32: 52–59
© Springer-Verlag 1997
O
OR
R II G
G II N
NA
AL
L PA
PA P
PE
ER
R
T. C. Harrington · D. L. McNew
Self-fertility and uni-directional mating-type switching
in Ceratocystis coerulescens, a filamentous ascomycete
Received: 6 July 1996 / 25 March 1997
Abstract Individual perithecia from selfings of most
Ceratocystis species produce both self-fertile and selfsterile progeny, apparently due to uni-directional matingtype switching. In C. coerulescens, male-only mutants of
otherwise hermaphroditic and self-fertile strains were selfsterile and were used in crossings to demonstrate that this
species has two mating-types. Only MAT-2 strains are
capable of selfing, and half of the progeny from a MAT-2
selfing are MAT-1. Male-only, MAT-2 mutants are selfsterile and cross only with MAT-1 strains. Similarly, selffertile strains generally cross with only MAT-1 strains.
MAT-1 strains only cross with MAT-2 strains and never self.
It is hypothesized that the switch in mating-type during
selfing is associated with a deletion of the MAT-2 gene.
Key words Homothallism · Mating-type switching ·
Hermaphroditism · Ceratocystis
Introduction
Sexual compatibility in heterothallic ascomycetes is determined by two alleles (idiomorphs) at a single mating-type
locus; two strains must be of opposite mating-type to produce ascospores (Glass and Kuldau 1992). Heterothallism
encourages outbreeding, which has obvious advantages,
but self-fertile species occur in many groups of ascomycetes. Various forms of homothallism are known (Beatty
et al. 1994), and in yeasts bi-directional mating-type
switching allows self-fertility in otherwise heterothallic
species (Herskowitz 1989). It has been proposed (Mathieson 1952; Perkins 1987; Fujii and Uhm 1988) that unidirectional mating-type switching allows for selfing in
T. C. Harrington (½) · D. L. McNew
Department of Plant Pathology, Iowa State University,
Ames, IA 50011, USA
Communicated by B. G. Turgeon
some filamentous ascomycetes. Although a switch in the
expression of mating-type is seen in these fungi, it is not
clear if a physical movement of mating-type genes is involved. It is also not clear if the expressed mating-types
of the respective self-fertile and self-sterile progeny are
homologs of the mating-type genes in other strictly heterothallic species of ascomycetes.
Sclerotinia trifoliorum and Chromocrea spinulosa show
a 1:1 segregation of self-fertile and self-sterile progeny in
perithecia from selfings or crosses (Mathieson 1952; Uhm
and Fujii 1983a, b). In tetrad analyses of selfings or crosses,
half of the ascospores in an ascus are large and give rise to
self-fertile colonies, and the other ascospores are small and
give rise to self-sterile colonies. Spore size appears to be
a pleiotropic character associated with self-fertility. The
self-fertile and self-sterile phenotypes have been referred
to as differing mating-types. The self-fertile mating-type
can switch to the self-sterile mating-type, but the reverse
mutation does not occur. A small percentage of the singleascospore progeny of Botryotinia fuckeliana may also self
and give rise to progeny of two mating-types (Faretra and
Pollastro 1996). A similar but more complicated form of
mating-type switching may occur in Glomerella cingulata
(Chilton and Wheeler 1949; Wheeler 1950).
Segregation for self-fertility and self-sterility is also
seen in many species of Ceratocystis sensu stricto (exclusive of Ophiostoma species such as O. ulmi), including
Ceratocystis fimbriata, C. albofundus, C. coerulescens and
C. coerulescens f. douglasii (Andrus and Harter 1937;
Olson 1949; Bakshi 1951; Davidson 1953; De Beer 1994).
In C. fimbriata, the ratio of self-fertile to self-sterile progeny in a single perithecium varies greatly (Olson 1949;
Webster and Butler 1967). Similar to the species with unidirectional mating-type switching, self-sterile progeny
from a selfing of C. fimbriata do not produce perithecia
when paired among themselves, but they can cross with
self-fertile progeny (Webster 1967). Ceratocystis eucalypti, C. fagacearum, C. radicicola and C. paradoxa are
reported to be strictly heterothallic (no self-fertility seen),
with each single-ascospore strain of one of two matingtypes (Dade 1928; Hepting et al. 1952; Kile et al. 1996;
53
El-Ani et al. 1957), typical of the mating system in other
filamentous ascomycetes. We (unpublished), however,
have found that progeny from a self-fertile isolate of
C. paradoxa from coconut palm in Brazil segregate for
self-fertility and self-sterility.
We hypothesized that species of Ceratocystis expressing self-fertility have a uni-directional mating-type switching mechanism analogous to that in S. trifoliorum and
C. spinulosa. Perithecia of Ceratocystis species show
adaptations for insect dispersal (Harrington 1993), and because the ascus wall dissolves while still in the perithecium we have been unable to conduct tetrad analyses on
these fungi. However, single-ascospore progeny of a single perithecium can be easily recovered from sticky ascospore masses at the tip of the perithecial neck. We (Harrington and McNew 1995) recovered both self-sterile and
self-fertile progeny from perithecia of several species of
Ceratocystis, including C. coerulescens, a fungus known
to cause blue stain in conifer sapwood (Harrington et al.
1996). This species proved particularly amenable to mating studies. We used self-sterile, male-only mutants of
otherwise hermaphroditic and self-fertile strains to demonstrate that this species has two mating-types, that only
MAT-2 strains are capable of selfing, and that half of the
progeny from a MAT-2 selfing are MAT-1.
Materials and methods
Strains. The isolates of C. coerulescens (species A) studied were
from blue-stained wood of Pinus species in England (Harrington
et al. 1996). The genotypic and phenotypic characters of these iso-
lates and other laboratory derived strains are given in Table 1. Isolate C798 was self-fertile and produced an abundant aerial mycelium and protoperithecia. Isolate C490 was a self-sterile subculture
from C798 that produced little aerial mycelium and few protoperithecia (designated the “flat” phenotype). Isolates C487, C488, C489
and C793 were also self-sterile but produced the typical aerial mycelium and abundant protoperithecia (designated the “aerial” phenotype). A cycloheximide-resistant mutant (C488R) was generated by
exposure to UV light. A conidial suspension was filtered through a
double layer of sterile cheesecloth, and 10 ml of the suspension was
dispersed into 50-mm-diameter Petri plates, exposed for 40 s to UV
light, and then spread onto MEA plates. Plates were incubated for
24 h in the dark and then overlaid with molten MEA with cycloheximide added at a concentration of 0.1 µg/ml. Developing colonies
were recovered, and MEA with 0.1 µg/ml cycloheximide was used
to evaluate resistance (growth rate as in MEA) or sensitivity (no
growth).
Crosses. Crosses were designated by a “DM” number, and singleascospore strains from such crosses were randomly assigned an “s”
number (e.g., DM44-s27 was the 27th single-ascospore strain from
DM44, a cross of C488 × C490). Cultures were grown and pairings
were conducted at room temperature and lighting. In most pairings,
recipient and spermatizing cultures were distinguished. Recipient
cultures were grown for 7–10 days on 1.5% malt extract with 2.0%
agar (MEA) in 95-mm-diameter plastic Petri dishes. In some cases,
an autoclaved 3-cm-long twig of Pinus strobus was placed in the recipient plate before the addition of molten MEA (Harrington 1992).
Spermatizing cultures were grown on MYEA (2.0% malt extract,
0.2% yeast extract, and 1.5% agar), which is more suited than MEA
for conidium production. The spermatizing cultures were flooded
with sterile, de-ionized water. Conidia, conidiophores and mycelial
fragments were loosened by lightly scraping with a spatula, and 1 or
2 ml of this suspension was added to each recipient culture. The recipient plate was swirled slightly to disperse the inoculum. Plates
were incubated at room temperature and inspected periodically for
perithecia and ascospore production, which was generally evident
within 2 weeks, but plates were observed periodically for up to
1 month. Other pairings, particularly those involving two flat strains
Table 1 Isolates and strains of C. coerulescens with their genotypic and phenotypic characters
Strain no.
Origin
Mating type
Mycelial
phenotype
Ability to self
Cycloheximide
resistance
C487
C488
C488R
C489
C490
C793
C795
C798
DM7-s1
DM7-s73
DM44-s27
DM77-s26
Field isolate
Field isolate
Mutant of C488
Field isolate
C798 subculture
Field isolate
Field isolate
Field isolate
C489×C490
C489×C490
C488×C490
C488×C490 progeny
backcrossed to C490
C488R×C490
C488R×C490
C488R×C490
C488×DM82-s52R
C488×C490 progeny
backcrossed to C490
C488×C490 progeny
backcrossed to C490
C488×DM82-s5R
C488R×DM82-s20
C488R× DM82-s20
C488×DM93-s62R
MAT-1
MAT-1
MAT-1
MAT-1
MAT-2
MAT-1
MAT-2
MAT-2
MAT-2
MAT-2
MAT-2
MAT-1
Aerial
Aerial
Aerial
Aerial
Flat
Aerial
Aerial
Aerial
Flat
Flat
Flat
Aerial
Self-sterile
Self-sterile
Self-sterile
Self-sterile
Self-sterile
Self-sterile
Self-fertile
Self-fertile
Self-sterile
Self-sterile
Self-sterile
Self-sterile
CyhS
CyhS
CyhR
CyhS
CyhS
CyhS
CyhS
CyhS
CyhS
CyhS
CyhS
CyhS
MAT-2
MAT-2
MAT-2
MAT-2
MAT-1
Flat
Aerial
Flat
Flat
Aerial
Self-sterile
Self-fertile
Self-sterile
Self-sterile
Self-sterile
CyhR
CyhS
CyhR
CyhS
CyhS
MAT-1
Aerial
Self-sterile
CyhS
MAT-2
MAT-2
MAT-2
MAT-2
Flat
Aerial
Aerial
Flat
Self-sterile
Self-fertile
Self-fertile
Self-sterile
CyhR
CyhS
CyhS
CyhS
DM82-s5R
DM82-s20
DM82-s52R
DM90-s28
DM92-s39
DM92-s48
DM93-s62R
DM102.P4-s9
DM102.P4-s13
DM183-s51
54
(strains with less aerial mycelium generally mated poorly), were conducted without regard to which strain served as the recipient or spermatizing partner. In these cases, spermatizing suspensions of two
strains were mixed in equal volumes, and about 1 ml of this suspension was added to a fresh plate of MEA or twig medium and swirled.
Self-fertility of single-ascospore strains was evaluated by incubating on MEA at room temperature and lighting. Perithecia production
was noted at 3 weeks. Aerial and flat phenotypes of strains were
also distinguished on MEA.
Isolation of ascospores. Ascospores of Ceratocystis species do not
disperse well in water (Whitney and Blauel 1972) but will disperse
in oil. Single-ascospore strains were obtained by collecting a mass
of exuded ascospores from the tip of a perithecial neck and dispersing in about 40 µl of Isopar M (a light sulfur oil, Exxon Corporation)
in a deep-well glass slide. Generally, about 5 µl of this suspension
was diluted in about 40 µl of Isopar M, and a small amount of this
spore suspension was picked up with a wire loop and streaked over
a 1.5% malt-agar plate. This plate was incubated at room temperature for 48 h, after which individual germlings were excised and
transferred to 1.5% malt-agar. Dilute spore suspensions proved to be
important in recovering representative populations of germlings,
without bias for speed of germination or growth rate.
Results
Self-fertile and self-sterile progeny from selfings
Among seven wild isolates of C. coerulescens examined,
only C795 and C798 were self-fertile (see Table 1). Among
the single-ascospore progeny recovered from a single perithecium of C798, 75 were self-fertile and 72 were selfsterile, closely fitting a 1:1 segregation ratio. All progeny
of C798 retained the aerial mycelial phenotype of the parent isolate.
Self-fertile strains generated from crosses of self-sterile
strains also produced self-fertile and self-sterile progeny.
The self-sterile isolate C490, a subculture of C798 that had
the flat mycelial phenotype, crossed with self-sterile isolates with the aerial phenotype, and some of the aerial phenotype progeny from such crosses were self-fertile. A total of 538 single-ascospore progeny were recovered from
selfings of self-fertile strains generated from ten crosses
between C488 × C490 or C489 × C490. All of the progeny
from these selfings were of the aerial phenotype, like the
parent strain, and 383 were self-fertile. The ratio of selffertile to self-sterile progeny was significantly greater than
that expected for a 1:1 segregation. We typically found this
deviation from 1:1 only in progeny recovered from dense
lawns of ascospores, apparently because of bias in sampling only the earliest-germinating progeny. When more
dilute ascospore suspensions were used and germlings harvested later, 1:1 segregation for self-fertility and selfsterility was seen in selfings of C. coerulescens, which
would be expected if selfing is due to a uni-directional
mating-type switch.
Two mating-types among self-sterile strains
Four self-sterile isolates of C. coerulescens showed evidence of two mating-types when paired among themselves
Table 2 Ascospore production in crosses between self-sterile isolates of C. coerulescens
Spermatizing
isolates
C490
C487
C488
C489
a
Recipient isolates
C490
MAT-2
C487
MAT-1
C488
MAT-1
C489
MAT-1
0/3a
3/3
1/3
0/3
3/3
0/3
0/3
0/3
5/5
0/3
0/3
0/3
5/5
0/3
0/3
0/3
Ratio of crosses with ascospore masses produced/attempted crosses
(Table 2). Isolate C490 and the self-fertile C798, from
which C490 was subcultured, were arbitrarily designated
MAT-2, while C487, C488 and C489 were designated
MAT-1. When used as recipients, the MAT-1 isolates consistently (13 positive out of 13 attempts) produced ascospores when spermatized by C490. However, C490 as a recipient produced fewer protoperithecia and perithecia than
the other strains, and ascospores were produced in only
4 of 9 cases where it was spermatized by MAT-1 isolates.
Because of its poor function as a female, we refer to
C490 as a male strain, while C798, C487, C488 and C489
were designated as hermaphrodites. In contrast to the other
isolates, C490 had a “flat” mycelial morphology. This phenotype had few protoperithecia, and these often took up to
3 weeks to develop, while the hermaphroditic isolates
(“aerial” phenotype) produced protoperithecia in less than
1 week. Also, the flat phenotype produced fewer conidiophores and less aerial mycelium, especially at the margins
of the plate, whereas the aerial phenotype produced many
conidiophores and an aerial mycelium spread uniformly
across the plate.
Two mating-types were identified among the self-sterile
progeny recovered from a single perithecium of the cross
C489 × C490 (designated DM7). Twelve of the DM7 progeny had the aerial phenotype and were self-fertile, six had
the aerial phenotype but were self-sterile, and 52 had the
flat phenotype. All of the flat-phenotype progeny were
self-sterile. The six aerial, self-sterile progeny and 12 randomly selected flat progeny were paired in all combinations, with each strain used both as a recipient and as a
spermatizing strain (Table 3). Six of the flat progeny were
compatible with the self-sterile, aerial progeny and with
MAT-1 C489, and these progeny were designated MAT-2.
The other six strains with the flat phenotype mated with
only the flat, MAT-2 progeny, and were thus identified as
MAT-1. All of the self-sterile, aerial progeny were MAT-1.
Both mating-types were identified among progeny recovered from crosses between flat strains of opposite
mating-type. Two MAT-1, flat-phenotype strains from the
cross C488 × C490 (DM44) were crossed with C490 or
DM44-s27, both of which were MAT-2, flat. Each of 115
progeny recovered from three crosses were of the flat phenotype, and none was self-fertile. Based on pairings with
MAT-1 (C488) and MAT-2 (DM7s-1) tester strains, 56 of
55
Table 3 Perithecium (P) and ascospore (A) production in crosses between aerial and flat, self-sterile progeny from the cross C489 (MAT-1,
aerial)×C490 (MAT-2, flat) of C. coerulescens
Spermatizing
strains
Recipient strains
MAT-1, aerial
MAT-1, flat
MAT-2, flat
13
16
21
28
60
69
2
7
15
34
35
44
1
6
22
39
57
73
13
16
21
28
60
69
–a
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
2
7
15
34
35
44
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
A
A
A
A
P
–
A
A
A
A
–
–
–
A
A
P
A
A
A
A
A
A
–
A
A
A
A
A
–
–
P
–
–
P
–
1
6
22
39
57
73
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
P
A
A
A
A
A
A
A
A
A
A
A
A
A
P
A
A
A
A
A
A
A
A
A
–
P
–
A
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
a
No perithecia (–), sterile perithecia (P), or perithecia with ascospores (A)
the progeny were MAT-1 and 59 were MAT-2, thus showing a 1:1 segregation (P = 0.05) for mating-type.
Flat phenotype and infertility
The aerial phenotype was strongly associated with reproductive ability or fertility in crosses, and appeared to be
essential for selfings to occur. Relatively few perithecia,
usually less than ten per plate, developed in the positive
pairings between flat-phenotype isolates of opposite
mating-type, and in some cases mature (full-sized) perithecia developed but no ascospore masses were exuded.
The flat MAT-2 strains were particularly poor as recipient
strains. Strain DM7-s73 served only as a male when
crossed with flat MAT-1 strains; the reciprocal crosses did
not result in perithecia (Table 3). In contrast, more than
100 perithecia with ascospores formed in each of the
36 plates where the aerial progeny served as the recipient
strain in compatible pairings (Table 3).
Only strains with the aerial phenotype were self-fertile,
and we were unable to generate self-fertile progeny from
crosses of flat strains of opposite mating-type. Each of
1360 progeny from 36 crosses between MAT-1, flat strains
(DM7 or DM44 progeny) and MAT-2, flat strains (C490)
was of the flat phenotype. Only 7 of these 1360 progeny
produced perithecia through selfing, and only 1–3 perithecia formed in the plates of the seven exceptional progeny.
Ascospore masses were seen on the top of six of these perithecia, and a total of 250 single-ascospore progeny were
recovered from four of them. None of these 250 progeny
was of the aerial phenotype, and only one produced perithecia. This strain produced two perithecia, one with no
ascospores and another with an ascospore mass and, when
this mass was transferred to another plate, no perithecia
were produced on the mycelium that developed. Thus,
progeny from crosses of flat strains were of the flat phenotype and essentially self-sterile. In the rare cases where
perithecia did form without pairing, these perithecia were
very few in number and often produced few viable ascospores. However, these flat isolates did form normal perithecia with ascospores, however, when paired with strains
of the opposite mating-type.
MAT-2, flat × MAT-1, aerial pairings
Segregation for mycelial phenotype and the ability to self
was examined in pairings between MAT-2, flat strains
(C490, as well as DM93-s62R and DM183-s51, from
crosses similar to DM7) with C488 and other MAT-1, aerial strains. Single-ascospore progeny were recovered from
seven perithecia. Half of the 661 progeny examined had
the aerial phenotype and half had the flat phenotype
(Table 4), showing that this character segregates as a single gene. Of the aerial phenotypes, half were self-fertile
and half were self-sterile. All of the flat progeny were selfsterile. The 1:1:2 segregation ratio (Table 4) is that expected for two factors (the MAT-2 mating-type and the aerial phenotype) required for self-fertility, and these factors
are unlinked.
56
Table 4 Segregation of self-fertility and mycelial phenotype in
single-ascospore progeny from individual perithecia from crosses of
self-sterile strains (MAT-1, aerial and MAT-2, flat) of C. coerulescens.
Number of Progenya
Parents
Aerial
MAT-1
Flat
MAT-2
DM92-s48
DM92-s48
DM92-s39
DM77-s26
C793
C488
C488
C490
C490
C490
C490
C490
DM93-s62R
DM183-s51
Total
Aerial
Aerial
Flat
self-fertile self-sterile self-sterile
21
28
26
22
24
17
28
25
25
18
21
29
23
24
53
48
56
34
44
52
43
166
165
330
sensitive to cycloheximide (DM90s-28), and all of the
progeny were self-sterile and showed sensitivity to cycloheximide. Ninety nine of the one hundred progeny showed
the flat phenotype of the MAT-2 parent, and one of the progeny showed an ambiguous mycelial phenotype. Of the 56
progeny for which mating-type was determined, 26 were
MAT-1 and 30 were MAT-2, which fits (P = 0.05) the expected 1:1 segregation for mating-type. The fact that the
progeny differed only in mating-type suggests that the aerial, MAT-1 parent induced selfing (mating-type switching)
of the flat, MAT-2 parent.
Mating type of self-fertile strains
a
Aerial:flat phenotypes and aerial, self-fertile:aerial, self-sterile
segregated in 1:1 ratios among progeny of each cross (chi-square,
P = 0.05)
Segregation for mycelial phenotype and mating-type
was examined in 96 single-ascospore progeny recovered
from a cross between C490 (flat, MAT-2) and DM92-s48
(aerial, MAT-1). Based on mycelial morphology on MEA
and pairings of the flat phenotypes with tester strains of
each mating-type, 28 of those 96 progeny were aerial, selffertile (presumably MAT-2); 25 were aerial, self-sterile
(presumably MAT-1); 21 were flat, MAT-1; and 22 were
flat, MAT-2. This segregation closely (P = 0.05) fits the
1:1:1:1 ratio expected for independent segregation of mycelial phenotype and mating-type.
Two crosses similar to the above were conducted with
either the MAT-1, aerial or the MAT-2, flat parent carrying
cycloheximide resistance. Analyses of the progeny from
each of the two crosses showed independent assortment of
mating-type, mycelial phenotype, and cycloheximide resistance (Table 5).
In the above crosses, independent assortment of markers from each parent was seen in the progeny, but in one
case we observed an induction of selfing in a flat, MAT-2
strain. A cycloheximide-resistant, MAT-1, aerial strain
(C488R) was crossed with a MAT-2, flat strain that was
We compared the ability of self-fertile strains to function
as MAT-2 or MAT-1 in crosses. The MAT-2 function of
13 self-fertile strains (self-fertile progeny from the cross
DM180, C488R × C490) was tested by spermatizing a
MAT-1, aerial strain with resistance to cycloheximide
(C488R) on cycloheximide medium. Because only the
MAT-1 recipient strain had resistance to cycloheximide, the
spermatizing strain could not self on the cycloheximide
medium. Crossing did occur in each of the 13 pairings because both cycloheximide-resistant and -sensitive progeny
were recovered from each of the perithecia sampled. All
287 of the recovered progeny from 13 perithecia (one perithecium from each cross) had the aerial phenotype of the
parents, 68 were self-fertile and cycloheximide-resistant,
81 were self-fertile and cycloheximide-sensitive, 73 were
self-sterile and cycloheximide-resistant, and 65 were selfsterile and cycloheximide-sensitive. This 1:1:1:1 segregation ratio (P = 0.05) indicates that self-fertility (matingtype) and cycloheximide resistance segregate independently, and that the self-fertile strain acted as a MAT-2 in
each of the crosses.
In order to determine if self-fertile strains could also
function as MAT-1 in a cross, 12 of the same self-fertile
strains (from DM180) were used to spermatize a flat,
MAT-2, cycloheximide-resistant strain (DM93s-62R)
growing on cycloheximide medium. As in the preceding
crosses, the recipient strain had cycloheximide resistance
while the spermatizing strains were sensitive, so the selffertile strains could not self on the cycloheximide medium.
Table 5 Segregation of self-fertility, mycelial phenotype, mating-type and cycloheximide resistance in single-ascospore strains from crosses
of self-sterile isolates of C. coerulescens of opposite mating-type with differing mycelial phenotype and cycloheximide resistance
Aerial phenotypea
Parents
Flat phenotype
Number
self-fertile
Number
MAT-1
Number
MAT-1
Number
MAT-2
MAT-1, aerial
MAT-2, flat
CyhR
CyhS
CyhR
CyhS
CyhR
CyhS
CyhR
CyhS
C488
C488R
DM93-s62R
C490
16
9
15
14
10
8
14
18
8
12
15
10
9
11
13
15
a
In each cross, segregation for the three characters (cycloheximide resistance, mycelial phenotype, and mating-type) occurred independently and did not differ significantly (P = 0.05) from the expected 1:1:1:1:1:1:1:1
57
None of these pairings resulted in perithecia, thus indicating that the self-fertile strains could not function as
MAT-1.
Further attempts to cross self-fertile strains and flat,
(self-sterile) MAT-2 strains utilized the above flat, MAT-2,
cycloheximide-resistant strain (DM93s-62R) growing on
cycloheximide medium and one of two self-fertile, cycloheximide-sensitive strains (DM102.P4s-9 or DM102.P4s13). In one such pairing, a few perithecia were produced,
but the ascospore masses that developed on the few perithecia were small, and a low germination rate of the ascospores was seen. Analysis of 39 recovered progeny showed
that 17 had the aerial phenotype, and 11 of these were
self-fertile. Thirteen of the twenty two flat progeny were
MAT-1 and nine were MAT-2. Of 39 progeny recovered, 19
were resistant to cycloheximide and 20 were sensitive, and
resistance was found in each of the classes noted above.
Thus, a self-fertile strain was able to spermatize a MAT-2,
flat strain in 1 out of the 14 attempted crosses, but in this
exceptional case few perithecia and viable ascospores were
produced. Analyses of the progeny showed independent
assortment of mating-type, mycelial phenotype, and cycloheximide resistance, confirming that a cross had occurred.
Lastly, 15 self-fertile and 15 self-sterile strains sensitive to cycloheximide were recovered from a selfing of isolate C795, and these were used as spermatizing strains
against cycloheximide-resistant strains on cycloheximide
medium. Again, the 15 self-sterile strains functioned only
as MAT-1 in crosses, and the 15 self-fertile strains functioned well as MAT-2 parents but not as MAT-1 parents in
crosses. When an aerial, MAT-1 strain (C488R) was used
as the recipient, no perithecia were produced when the selfsterile strains were used as spermatizing strains, but perithecia with ascospores were produced in 45 out of 45 of
the attempted crosses with the self-fertile strains. When a
flat, MAT-2 strain (DM93-62R) was used as the recipient,
28 out of 45 crosses with the self-sterile strains resulted in
perithecia with ascospores, but only 1 of 45 crosses with
the self-fertile strains resulted in a perithecium with an ascospore mass. Fifty four progeny from this perithecium
were recovered, and analyses showed that mating-type,
mycelial phenotype, and cycloheximide resistance each
segregated independently and in a 1:1 ratio. Thus, in this
one case, a self-fertile strain functioned as a MAT-1 in a
cross, but self-fertile strains generally function only as
MAT-2 in crosses.
Discussion
Two mating-types can be identified in the heterothallic species C. coerulescens, but one of the mating-types, arbitrarily designated MAT-2, is typically self-fertile and gives
rise to both MAT-2 (self-fertile) and MAT-1 (self-sterile)
progeny. The switch in expression of mating-type in selffertile strains is uni-directional, i.e., MAT-2 strains can
switch to MAT-1 and self, but MAT-1 progeny cannot switch
to MAT-2 and thus remain self-sterile. This provides the
strongest evidence to-date that unidirectional mating-type
switching occurs in ascomycetes.
The change in expression of mating-type does not appear to occur in the vegetative phase of the fungus. We
have observed that single-conidium progeny of MAT-2 and
MAT-1 strains of C. coerulescens always retain the matingtype of the parent strain (unpublished); Webster and Butler (1967) observed the same phenomenon in C. fimbriata.
Generally, self-fertile strains did not function as MAT-1
when used as spermatizing strains, and only MAT-2 strains
capable of protoperithecium production (the aerial phenotype) were able to self. Thus, the switch in mating-type
may occur in the protoperithecium, perhaps before dikaryon formation.
We assume that the two mating-types in Ceratocystis
are idiomorphs at a single locus, which is typical for other
heterothallic ascomycetes (Glass and Kuldau 1992). In
other work, we (Harrington and McNew 1995) have found
that C. coerulescens strains are partially interfertile with
strains of other “self-fertile” species, a strictly heterothallic species (C. eucalypti), and putatively asexual species
(e.g., Chalara neocalidoniae) in the C. coerulescens complex (Harrington 1996), but only if the paired strains are
of opposite mating-type. Because the two mating-types in
species of Ceratocystis with self-fertility are functionally
compatible with the two mating-types in other members of
this genus, we believe the two mating-types are homologous; hence the designation of MAT-1 and MAT-2 for the
mating-types of the self-fertile species.
Using the polymerase chain reaction and degenerative
primers designed for the High Mobility Group (HMG) box
of the mt a-1 sequence in Neurospora crassa (Arie et al.
1997), we (Strydom et al., unpublished data) amplified a
product of approximately 300 bp from DNA of MAT-2, but
not MAT-1, strains of C. eucalypti, Chalara australis and
Ch. neocalidoniae. Internal primers specific to the C. eucalypti MAT-2 sequence were designed, and these also specifically amplified a portion of the MAT-2 idiomorph from
DNA of C. eucalypti and the closely related Chalara species. Unfortunately, the degenerative primers and the primers designed from the C. eucalypti MAT-2 sequence did
not amplify a single 300-bp product from DNA of the
C. coerulescens MAT-2 mating-type. However, the C. eucalypti primers amplified the expected fragment from the
DNA of self-fertile isolates and other MAT-2 isolates of
C. virescens, a species closely related to C. eucalypti that
also undergoes mating-type switching (Harrington and
McNew 1995). No such fragment was produced with the
DNA from a MAT-1, self-sterile isolate of C. virescens. We
also recovered self-fertile and self-sterile progeny from the
selfing of a self-fertile isolate of C. virescens, and the
C. eucalypti primers amplified the expected fragment from
DNA of the ten self-fertile progeny, but no fragment was
amplified from DNA of the ten self-sterile progeny. Thus,
we have demonstrated with these primers that at least a
portion of the MAT-2 idiomorph is deleted in a selfing
event.
True homothallic species of other genera of filamentous
ascomycetes generally carry the idiomorphs of both
58
mating-types (Beatty et al. 1994; Turgeon et al. 1995),
and we hypothesize that MAT-2 strains of ascomycetes
with unidirectional mating-type switching carry both the
MAT-2 and MAT-1 idiomorphs, while MAT-1 strains carry
only the MAT-1 idiomorph. The preliminary findings
using MAT-2-specific primers, along with the results from
numerous pairings, support this hypothesis. Selfings of
MAT-2 strains produce progeny of both mating-types, and
self-fertile strains can cross with both MAT-1 and MAT-2
strains, although the latter cross is apparently rare. In contrast, MAT-1 strains do not function as MAT-2 in crosses
and do not self, so the change in expression of mating-type
appears irreversible, perhaps due to a simple deletion of
the MAT-2 idiomorph.
The presence of MAT-2 and MAT-1 in self-fertile strains
is not likely to be due to diploid nuclei or dikaryons in ascospores. Analyses of 98 isolates of C. coerulescens and
related species for ten isozyme loci showed no evidence of
heterozygosity (Harrington et al. 1996), and no heterozygosity was seen in analyses of C. coerulescens progeny
from crosses of parents differing in isozyme electromorphs
(unpublished data). Also, microscopic examination has
shown that ascospores of this species are uninucleate (unpublished data). Thus, the species is apparently haploid,
and single-ascospore progeny are homokaryotic. Aneuploidy is a possibility, however, and deletion of the chromosome, or part of the chromosome, with the MAT-2 idiomorph may be responsible for the change in mating-type
expression.
It is noteworthy that in the other filamentous ascomycetes with unidirectional mating-type switching, selffertile strains arise from ascospores that are larger than the
ascospores that give rise to self-sterile strains (Mathieson
1952; Ulm and Fujii 1983a), perhaps due to higher DNA
content in the nuclei of the self-fertile strains. We have not
seen evidence for two different sizes of ascospores in
Ceratocystis, but we (Harrington and McNew 1995) have
seen differences in growth rate between the two matingtypes, with MAT-2 strains growing significantly faster than
MAT-1 strains in all the tested species of Ceratocystis with
mating-type switching. Like the smaller ascospore size in
the other genera, the slower growth rate seen in self-sterile
progeny of Ceratocystis species is a pleiotropic character
associated with mating-type and may be due to a loss of
genetic material or other chromosome rearrangement.
It is not clear if the MAT-1 idiomorph is present as a silent copy in self-fertile strains or if a translocation of this
copy is necessary for selfing to take place, in a manner similar to that in yeasts with mating-type switching (Herskowitz 1989). If physical movement of DNA does take place,
it is apparently irreversible, unlike that in yeasts. In our
proposed model, idiomorphs of both mating-types may be
in functional locations in self-fertile strains, and the apparent shift in expression of mating-type to MAT-1 is merely
a loss in function of the gene(s) at the MAT-2 idiomorph,
perhaps by deletion. Such a deletion might be expected to
have pleiotropic effects, such as the slower growth rate of
MAT-1 strains when compared to MAT-2 strains (Harrington and McNew 1995). Strains of heterothallic filamentous
ascomycetes that have been transformed with a gene of opposite mating-type are capable of selfing, i.e., producing
pseudothecia or perithecia but they are sterile (no asci or
ascospores) (Picard et al. 1991; Turgeon et al. 1993). The
transgene confers the ability to form pseudothecia but not
asci and ascospores in crosses, if the resident MAT gene
in the transgenic strain is suppressed (Turgeon et al. 1993;
Wirsel et al. 1996). Self-fertile strains of C. coerulescens
can self but generally only function as MAT-2 in crosses,
perhaps because they remain functionally MAT-2 until the
MAT-2 gene is deleted, sometime after protoperithecium
production. The precise stage of the switch in mating-type
is not known, but it may be before the formation of the dikaryon. The expression of the MAT-1 idiomorph would
then allow for complementation with an “unswitched” nucleus, so that the sexual cycle proceeds, but half of the
progeny only express the MAT-1 mating-type and cannot
revert to MAT-2.
Our findings with C. coerulescens are consistent with
the reports of Olson (1949), Webster (1967), and Webster
and Butler (1967) for C. fimbriata, in which self-sterile to
self-fertile progeny are recovered from a single-ascospore
mass, although the self-fertile progeny typically outnumber the self-sterile progeny in that species. In C. coerulescens, C. fimbriata and other Ceratocystis species, we have
seen some segregation ratios that are slightly skewed, generally in favor of self-fertile progeny, and the mechanism
for that skewness is not clear. In C. coerulescens, we
often saw skewed segregation for mating-type and other
characters when we recovered germlings from dense lawns
of ascospores, presumably because only hyphae of the earliest germinating ascospores could be recovered before
intermingling. In C. fimbriata, germination percentages of
ascospores are generally low, and self-sterile germlings
may have a lower viability or the ascospores may germinate slower than self-fertile ascospores, resulting in recovery of fewer self-sterile progeny. We (Harrington and
McNew 1995) have noted that colonies of MAT-2 strains
of C. fimbriata have a faster linear growth rate than colonies of MAT-1 strains. Alternatively, in the developing perithecia from selfings or crosses, the ascogenous system
may be a mixture of dikaryotic hyphae, some with MAT-1
and MAT-2 nuclei paired and others with only MAT-2 nuclei. Regardless of this, normal segregation ratios for
mating-type and other markers were generally seen in
C. coerulescens when sparse lawns of ascospores were
used to recover individual germlings, which we feel is better suited to unbiased recovery of fast- and slow-germinating progeny.
Although isolates of C. coerulescens and other Ceratocystis species are typically hermaphroditic, we have been
able to identify male mutants in many of these species. It
is not clear if such male mutants are common in nature or
are an artifact of laboratory culture. Male mutants were
earlier reported in C. fimbriata and C. fagacearum cultures
(Barnett 1953; Webster and Butler 1967), and loss of “femaleness” is apparently common in ascomycetes (Hansen
1938; Hansen and Snyder 1943; Notteghem and Silué
1992; Linders and Van der Aa 1995). In Ceratocystis these
59
mutants generally produce few protoperithecia, and in
C. coerulescens fewer conidiophores and less aerial mycelium are seen in the male (flat) mutants than in wild-type
(aerial) isolates. We rarely saw MAT-2 , flat isolates self,
and in the one time this was studied in detail it was apparently due to a mating-type switching event that was induced by pairing with a MAT-1 strain. Regardless of
mating-type, the flat phenotype is usually associated with
low fertility, particularly when serving as a female. These
mutations must impede the expression of many genes, but
the factor is not linked to the mating-type locus.
Whatever the mechanism for selfing, it is apparently
common in the genus Ceratocystis and appears similar
to that noted in a few other filamentous ascomycetes
(Mathieson 1952; Uhm and Fujii 1983a, b; Perkins 1987;
Fujii and Uhm 1988; Rehnstrom and Free 1993). Further
study is needed to determine the molecular mechanism of
mating-type switching in Ceratocystis and other filamentous ascomycetes. With cloned fragments of the MAT-2
idiomorph we will be able to more precisely follow the
events leading to changes in mating-type.
Acknowledgements The helpful comments of Roger Wise and Charlotte Bronson are greatly appreciated. Gillian Turgeon generously
provided the sequences of the degenerative primers used to amplify
the MAT-2 HMG box in C. eucalypti, and Corli Strydom assisted with
the PCR work. Journal Paper No. J-17002 of the Iowa Agriculture
and Home Economics Experiment Station, Ames, Iowa. Project
No. 3226, supported by Hatch Act and State of Iowa Funds.
References
Andrus C, Harter L (1933) Morphology of reproduction in Ceratostomella fimbriata. J Agric Res 46:1059–1078
Arie T, Christiansen S, Yoder O, Turgeon B (1997) Efficient cloning of ascomycete mating-type genes by PCR amplification of
the conserved MAT HMG box. Fungal Genet Biol 21:118–130
Bakshi B (1951) Studies on four species of Ceratocystis, with a discussion of fungi causing sapstain in Britain. Commonwealth
Mycol Inst, Mycol Papers 35:1–16
Barnett H (1953) A unisexual male culture of Chalara quercina.
Mycologia 45:450–457
Beatty N, Smith M, Glass N (1994) Molecular characterization of
mating-type loci in selected homothallic species of Neurospora,
Gelasinospora and Anixella. Mycol Res 98:1309–1316
Chilton S, Wheeler H (1949) Genetics of Glomerella. VII. Mutation
and segregation in plus cultures. Am J Bot 36:717–721
Dade H (1928) Ceratostomella paradoxa, the perfect stage of Thielaviopsis paradoxa (de Seynes) von Höhnel. Trans Br Mycol Soc
13: 184–193
Davidson R (1953) Two common lumber-staining fungi in the western United States. Mycologia 45: 579–586
De Beer C (1994) Ceratocystis fimbriata with special reference to
its occurrence as a pathogen of Acacia mearnsii in South Africa.
MS thesis, University of the Orange Free State, Bloemfontein,
South Africa
El-Ani S, Klotz L, Wilbur W (1957) Heterothallism, heterokaryosis,
and inheritance of brown perithecia in Ceratostomella radicicola. Mycologia 49:181–187
Faretra F, Pollastro S (1996) Genetic studies of the phytopathogenic fungus Botryotinia fuckeliana (Botrytis cinerea) by analysis of
ordered tetrads. Mycol Res 100:620–624
Fujii H, Uhm J (1988) Sclerotinia trifoliorum, cause of rots of Trifolium spp. Adv Plant Pathol 6:233–240
Glass N, Kuldau G (1992) Mating type and vegetative incompatibility in filamentous ascomycetes. Annu Rev Phytopathol 30:
201–224
Hansen H (1938) The dual phenomenon in imperfect fungi. Mycologia 30:442–455
Hansen H, Snyder W (1943) The dual phenomenon and sex in Hypomyces solani f. cucurbitae. Am J Bot 30:419–422
Harrington T (1992) LeptographiumIn: Singleton LL, Mihail JD,
Rush CM (eds) Methods for research on soilborne phytopathogenic fungi. APS Press, St. Paul, Minnesota, pp 129–133
Harrington T (1993) Biology and taxonomy of fungi associated with
bark beetles. In: Schowalter TD, Filip GM (eds) Beetle-Pathogen
Interactions in Conifer Forests. Academic Press, New York,
pp 37–58
Harrington T, McNew D (1995) Mating-type switching and selffertility in Ceratocystis (abstract). Inoculum 46:18
Harrington T, Steimel J, Wingfield M, Kile G (1996) Isozyme variation and species delimitation in the Ceratocystis coerulescens
complex. Mycologia 88:104–113
Hepting G, Toole E, Boyce J (1952) Sexuality in the oak wilt fungus. Phytopathology 42:438–442
Herskowitz I (1989) A regulatory hierarchy of cell specialization in
yeast. Nature 342:749–757
Hunt J (1956) Taxonomy of the genus Ceratocystis. Lloydia 19: 1–58
Kile G, Harrington T, Yuan Z, Dudzinski M, Old K (1996) Ceratocystis eucalypti sp. nov., a vascular-stain fungus from eucalypts
in Australia. Mycol Res 100:571–579
Linders E, Van der Aa H (1995) Taxonomy, sexuality and matingtypes of Diaporthe adunca. Mycol Res 99:1409–1416
Mathieson M (1952) Ascospore dimorphism and mating-type in
Chromocrea spinulosa (Fuckell) Petch n. comb. Ann Bot NS
16:449–467
Notteghem J, Silué D (1992) Distribution of the mating-type alleles
in Magnaporthe grisea populations pathogenic on rice. Phytopathology 82:421–424
Olson, E (1949) Genetics of Ceratostomella. 1. Strains in Ceratostomella fimbriata (Ell. & Hals.) Elliott from sweet potatoes. Phytopathology 39:548–561
Perkins D (1987) Mating-type switching in filamentous ascomycetes. Genetics 115:215–216
Picard M, Debuchy R, Coppin E (1991) Cloning the mating-types
of the heterothallic fungus Podospora anserina: developmental
features of haploid transformants carrying both mating-types.
Genetics 128:539–547
Rehnstrom A, Free S (1993) A simple method for the mating of Sclerotinia trifoliorum. Exp Mycol 17:236–239
Turgeon B, Christiansen S, Yoder O (1993) Mating-type genes in
Ascomycetes and their imperfect relatives. In: Reynolds DR,
Taylor JW (eds) The fungal holomorph: mitotic, meiotic and pleomorphic speciation in fungal systematics. CAB International,
Oxford, UK, pp 199–215
Turgeon B, Sharon A, Wirsel S, Yamaguchi K, Christiansen S,
Yoder O (1995) Structure and function of mating-type genes in
Cochliobolus spp. and asexual fungi. Can J Bot 73 (suppl 1):
S778–S783
Uhm J, Fujii H (1983a) Ascospore dimorphism in Sclerotinia trifoliorum and cultural characters of strains from different-sized
spores. Phytopathology 73:565–569
Uhm J, Fujii H (1983b) Heterothallism and mating-type mutation in
Sclerotinia trifoliorum. Phytopathology 73:569–572
Webster R (1967) The inheritance of sexuality, color and colony type
in Ceratocystis fimbriata. Mycologia 59:222–234
Webster R, Butler E (1967) The origin of self-sterile, cross-fertile
strains in Ceratocystis fimbriata. Mycologia 59:212–221
Wheeler H (1950) Genetics of Glomerella. VIII. A genetic basis for
the occurrence of minus mutants. Am J Bot 37:304–312
Whitney H, Blauel R (1972) Ascospore dispersion in Ceratocystis
spp. and Europhium clavigerum in conifer resin. Mycologia
64:410–414
Wirsel S, Turgeon B, Yoder O (1996) Deletion of the Cochliobolus
heterostrophus mating-type (MAT) locus promotes the function
of MAT transgenes. Curr Genet 29:241–249